Food Quality and Preference 50 (2016) 5764

Contents lists available at ScienceDirect

Food Quality and Preference

journal homepage: www.elsevier.com/locate/foodqual

Parental feeding practices, food neophobia, and child food preferences:

What combination of factors results in children eating a variety of foods?
Jill L. Kaar a,, Allison L.B. Shapiro b, Donna M. Fell c, Susan L. Johnson a
a

Department of Pediatrics, Colorado School of Medicine, University of Colorado Anschutz Medical Campus, United States
Department of Epidemiology, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, United States
c
Department of Nutrition, Metropolitan State University of Denver, United States
b

a r t i c l e

i n f o

Article history:
Received 14 July 2015
Received in revised form 12 January 2016
Accepted 14 January 2016
Available online 16 January 2016
Keywords:
Parental feeding practices
Child food neophobia
Food preferences
Eating behaviors
Food acceptance
Parentchild concordance

a b s t r a c t
Our aim was to describe food neophobia, parenting feeding practices, and concordance in food preferences between parentchild dyads using a cross-sectional on-line survey completed by parents of
preschoolers (35 y). Respondents (n = 210) included mothers (89%) who were predominantly white
(85%) and college educated (64%). Most children (mean age = 41.7 mo 14.9), were perceived to be of
a healthy weight (81%) and good eaters (60%). Parent (21.9 7.4) and child food neophobia
(30.4 8.8) correlated significantly, though modestly (r = 0.14, p = 0.04). The parent practice of offering
new foods to eat was inversely associated with child food neophobia (r = 0.40, p < .0001) and pressure
to eat (r = 0.13, p = 0.07). Parentchild dyads had >75% concordance in preferences for whole grains,
fruit, starchy vegetables, and dairy. Lowest concordance (3966%) was noted for sweetened beverages
and entrees. Discordance occurred when parents had never offered their children a food (e.g., vegetables),
and was uninfluenced by demographic factors. A childs food neophobia and overweight status was associated with the child having a lower consumption of vegetables. Parent reports of giving the child more
control of food-related decisions was associated with a higher number of healthy foods rated as liked by a
child. Similarities in parentchild food preferences may be related to food neophobia and, consequently,
the foods parents offer their children. Educating parents on the potential impact of feeding practices may
be important for early intervention efforts to improve childrens food acceptance.
2016 Elsevier Ltd. All rights reserved.

1. Introduction
Food preferences and eating patterns develop in early childhood
and remain relatively stable through adolescence (Northstone &
Emmett, 2008) and adulthood (Mikkila, Rasanen, Raitakari,
Pietinen, & Viikari, 2005). Therefore, early childhood may represent
a sensitive window of development for establishing good eating
practices and healthy food preferences that could potentially
impact an individuals lifelong health. Known factors impacting
the development of a childs food preferences include parental
characteristics such as parental feeding practices, parents own
childhood eating experiences, and caregivers food neophobia (fear
of trying new foods) (Birch, 1999; Fletcher, Branen, & Lawrence,
1997; Ventura & Birch, 2008; Vereecken, Rovner, & Maes, 2010),

Corresponding author at: Department of Pediatrics, University of Colorado

School of Medicine, Anschutz Medical Campus, Mail Stop F561, Academic Office 1,
12631 E. 17th Ave, Rm. L15-2615, Aurora, CO 80045, United States.
E-mail address: Jill.Kaar@ucdenver.edu (J.L. Kaar).
http://dx.doi.org/10.1016/j.foodqual.2016.01.006
0950-3293/ 2016 Elsevier Ltd. All rights reserved.

as well as a childs own food neophobia (Fletcher et al., 1997;

Ventura & Birch, 2008).
The influence of parents and their behaviors surrounding eating
and food choices (Bova & Arcidiacono, 2014) suggests that parents
play a key role in shaping a childs food preferences and eating patterns (Gregory, Paxton, & Brozovic, 2011). Parental feeding practices that influence childrens development of food preferences
and intake patterns include modeling, restriction of foods, and
pressuring a child to eat (Birch, 1999; Vereecken et al., 2010).
Specifically, pressuring a child to eat is negatively associated with
food acceptance (Birch & Marlin, 1982; Newman & Taylor, 1992),
may negatively impact overall diet quality (Scaglioni, Salvioni, &
Galimberti, 2008), and has been found to be a significant mediating
factor between maternal feeding beliefs and food neophobia
(Cassells, Magarey, Daniels, & Mallan, 2014). Alternatively, parental modeling of healthy eating has been shown to correlate with
lower rates of childrens picky eating (Gregory, Paxton, &
Brozovic, 2010). Furthermore, parents influence on a childs diet
via the foods they make available in the environment and the practices they use to offer foods to their children may also contribute to

58

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

a childs own development of food neophobia which in turn can

further impact the childs food preferences and healthy eating
choices (Boles et al., 2014).
Food neophobia has been linked to the development of food
preferences and in turn, limited diet variety (Howard, Mallan,
Byrne, Magarey, & Daniels, 2012). Food neophobia has also been
linked to both environmental and genetic (maladaptive trait)
determinants (Cooke, Haworth, & Wardle, 2007; Dovey, Staples,
Gibson, & Halford, 2008; Faith, Heo, Keller, & Pietrobelli, 2013).
The individual level characteristics of child food neophobia has
been shown to significantly shape a childs early food preferences
and dietary intake (Cassells et al., 2014; Fletcher et al., 1997;
Ventura & Birch, 2008; Laureati et al., 2015; Russell & Worsley,
2008; Johnson, Davies, Boles, Gavin, & Bellows, 2015), and also lessen a childs overall enjoyment of food which can in turn affect
food choices (Mustonen, Oerlemans, & Tuorila, 2012). Studies have
reported significant concordance rates between childrens and parents (mainly mothers) food neophobia (Galloway, Lee, & Birch,
2003), further implicating the potential genetic link between parental factors, such as parental food neophobia in the development
of child eating behaviors and preferences. In one study, mothers
who reported higher food neophobia scores were less likely to offer
healthy foods to their children (Cooke & Wardle, 2005; Skinner,
Carruth, Wendy, & Ziegler, 2002). Yet in another study, mothers
who scored high for food neophobia were found to use more
restriction in feeding and less monitoring (Tan & Holub, 2012). In
turn, these parental feeding practices have been associated with
food neophobia in children (Tan & Holub, 2012; Vereecken et al.,
2010). Children with food neophobia have been reported to consume a limited variety of foods (Cooke, Carnell, & Wardle, 2006;
Howard et al., 2012) and in general eat fewer servings of vegetables compared to children who do not have food neophobia
(Birch, 1999; Dovey et al., 2008; Galloway et al., 2003; Laureati
et al., 2015). Therefore, parental characteristics such as food neophobia, feeding practices, and the childs own food neophobia
could ultimately influence childrens weight status.
Parental pressure to eat and restriction, have been associated
with child weight (Boles, Reiter-Purtill, & Zeller, 2013; Brown &
Lee, 2011; Gregory et al., 2010; Haszard, Williams, Dawson,
Skidmore, & Taylor, 2013; Ventura & Worobey, 2013), suggesting
that in addition to contributing to the development of poor dietary
patterns, parent feeding practices may also predispose children to
weight issues early in life. Parents restriction in feeding has been
reported to be associated with larger perceived infant size while
pressuring to eat has been associated with smaller perceived infant
size (Brown & Lee, 2011). These findings suggest that the association between parental perceptions of childrens weight and their
child feeding practices begin early in life and may have important
implications for the development of childrens food preferences
and their dietary intakes. The potential pathway between parental
feeding practices and development of child food neophobia, and
food preferences, warrants investigation into which elements of
parental and child behavior within this pathway have the most significant impact on the outcome of child food preferences.
To date, the majority of studies examining the influences of parental feeding practices and childrens food acceptance have only
examined individual factors (e.g., pressuring to eat), without
simultaneously accounting for other potential factors such as parent food related history and traits, multiple parental feeding practices, child food neophobia, and parent and child food preferences.
To examine the relative contribution of parental characteristics
(including the heritability of food neophobia), feeding practices,
and child neophobia to child food preferences, this study aimed
to: (1) describe food neophobia and parental feeding practices in
parentchild dyads; (2) assess concordance between parentchild
food preferences; (3) examine the relationship between parental

feeding practices, child food neophobia with overall child food

preferences using a proposed theoretical model (Fig. 1); and (4)
identify characteristics (e.g., feeding practices, food neophobia)
associated with a child eating healthy foods including vegetables,
fruits, and core food items.
2. Methods
2.1. Participants
University email listservs (Metropolitan State University of
Denver, CO and University of Colorado Anschutz Medical Campus,
Aurora, CO) were used to recruit parents of preschoolers
(35 years). Participants included in this study (N = 210 parent
child dyads) were those parents who responded to recruitment
emails and consented to completing an online survey to assess
feeding practices and child food preferences. This study was
approved by the Colorado Multiple Institutional Review Board
(COMIRB) under the exempt category and therefore all participants
were deemed consented if they volunteered to submit the survey.
Study data were collected and managed using the REDCap
electronic data capture tool hosted at the University of Colorado
(Harris et al., 2009).
2.2. Measures
An online Research Electronic Data Capture (REDCap) (Harris
et al., 2009) survey was developed using several validated
instruments to collect information regarding variables of interest
(food neophobia, food preferences, parental feeding practices,
and demographic information).
Food preferences for 80 food items were assessed for both
adults and children, using a shortened version of Skinner et al. list
of foods (Skinner et al., 2002). To avoid over-burdening participants, a branching logic was used to assign 40 foods that represented food from all food groups: girls were assigned one set of
40 foods and boys were assigned another 40 foods. The number
of foods assessed (n = 40) was balanced between boys and girls
to ensure all food groups were accounted for using equal numbers
of food options (see food list in appendix). Specifically, child food
preferences were measured by asking parents to report their
childs liking for the foods. Each item was evaluated on a 5 point
Likert-like scale (1 = like and eat, 2 = like and do not eat, 3 = dislike
and eat, 4 = dislike and do not eat, and 5 = never tasted/I have
never offered this food to my child) (Skinner et al., 1998). By
design, the questionnaire randomly selected 40 food items for a
child depending on gender; hence the analysis was conducted separately for boys and girls. Adult and child food neophobia were
assessed using the adult and child versions of the food neophobia
scale (FNS; testretest reliability ranged from r = 0.820.91)
(Pliner, 1994; Pliner & Hobden, 1992). Parents responded by their
level of agreement (scale 15) to specific statements (e.g. My child
is constantly sampling new and different foods and If my child
doesnt know what is in a food, my child wont try it). These
instruments have been shown to have satisfactory testretest
and internal reliability. Possible scores ranged from 10 to 50, with
a higher score indicating greater food neophobia.
Parental feeding practices were measured using the Child
Feeding Questionnaire (CFQ-CO), an instrument that combines
the original CFQ by Birch and colleagues (2001; Cronbachs
alpha = 0.700.92) with three new constructs added for this study
(Birch et al., 2001). The constructs included pressure to eat (from
the original CFQ) and three added constructs, Child Autonomy
(giving the child control over food-related decisions), Parent
Offering Child New Foods, and Parent Urging of New Foods.

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

59

Fig. 1. Tested theoretical model of the development of child food preferences via parents traits (eating history and food neophobia), parental feeding practices, and childs
food neophobia.

Example items for the additional constructs included (1) Child

Autonomy: My child decides how much of each food goes on
his/her plate and My child is old enough to feed him/herself,
(2) Parent Offering Child New Foods: When I offer a new food, I
have my child eat at least one bite and I serve only the foods that
my child likes (reverse score), and (3) Parent Urging of New
Foods: I have my child try a new food before they eat sweet foods
and If my child refuses to eat a new food or a food they dont like, I
continue to offer that food to him/her. Each item for all constructs
was scored on a scale from 1 to 5 where 1 = Never and 5 = Always.
Scores from items were summed for each of the constructs and
averaged for a score that ranged from 1 to 5. A higher average score
for the promotion of Child Autonomy and Parental Offering of New
Foods constructs indicated a positive feeding practice, whereas a
higher score for Parental Urging of New Foods indicated a counterproductive practice. Parent eating history, which assessed parents
memories of eating as a child, was retrospectively assessed using
the Eating Then & Now Questionnaire (Branen & Fletcher, 1999).
This instrument links childhood food related experiences to the
development of an individuals relationship with food as an adult.
Parents respond to questions based on their memories of when
they were a child (e.g. I was allowed to choose which foods I
ate at meals and I served my own food.) Scores for this construct
ranged from 5 to 25, with a lower score indicating more autonomy
to make food/eating decisions as a child. Demographic details were
also included in the survey (including: race, ethnicity, education,
family income, employment status, parent self-reported height
and weight, frequency of family dinners, age of child, birth order
of child, and childs weight status). Parents rated (yes/no response)
their childs willingness to eat by asking the parent do you think
your child is a good eater, giving us the parents general perception of the amount or type of food their child eats. Information on
self-reported height and weight of parent completing the survey
was used to calculate parent body mass index.
2.3. Statistical analysis
All analyses were conducted in SAS v9.4 (SAS Institute, Cary,
NC). Distributions of the data were first examined using visual
inspection via histograms and measures of skewness. All variables
were found to be normally distributed. Parent and child demographics were compared between categories using either t-tests
for continuous variables or chi-square tests for categorical
variables. The correlations among continuous variables including
concordance between parentchild dyads and preferences for
specific food groups, were calculated using Pearson correlation
coefficients. Analysis of variance was used to examine potential

demographic factors that may have influenced agreement between

parentchild dyads. Due to the nature of the data collection
methods, all findings are reported separately for boys and girls.
Forward stepwise linear regression models were used to select
variables significantly associated with the number of foods a child
was reported to like and eat within each food group. Individual
foods were grouped into five food groups using the approach from
Howard et al. in which non-core foods represented energy dense,
little nutritional value foods (French fries, chicken nuggets, cookies) and core foods as healthy food choices (milk, yogurt, eggs, lean
meats). The remaining groups include vegetables, fruits, and
sweetened beverages (Howard et al., 2012). Variables included in
each of the forward steps were: Step (1) parent demographics:
race, education, income, employment status and Parent BMI; Step
(2) parent characteristics: food neophobia and eating history; Step
(3) parental feeding practices: promotion of Child Autonomy, pressure to eat, offering new foods and urging new foods; Step (4) child
demographics: age, birth order and weight status; and Step (5)
child characteristics: food neophobia, bringing forward only the
significant variables from the previous model. The final model
(step) included only significant variables from step 5 (Table 5).
For each step, R2 was reported and change in R2 between steps
was calculated using Partial F-test procedures and reported if
significant at the p < 0.05 level. The effect size was calculated to
illustrate the strength of association for the remaining variables
to each outcome and is presented in the final model.
3. Results
Participants (N = 210) were mostly mothers (89%), nonHispanic white (85%), and college educated (64%). Children of the
participants were an average age of 41.7 months (14.9 months)
and most (81%) were perceived by their parent to be of a healthy
weight and to be good eaters (60%; Table 1). Baseline characteristics of this sample did not differ between boys and girls, with the
exception that girls were more likely than boys to be categorized as
good eaters by their parents (67% vs 52%, p = 0.03).
3.1. Food neophobia and parental feeding practices
Parent 21.9 7.4 and child 30.4 8.8 food neophobia scores were
modestly correlated (r = 0.14, p = 0.04; Table 2). Parent eating
history scores (from the Parent Eating Then and Now Questionnaire)
ranged from 7 to 25 with an average score of 16.4 3.8. Parent eating
history and parental feeding practices scores did not differ by child
gender. The four parental feeding practices are reported according
to child gender in Table 2.

60

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

Table 1
Sample characteristics for both parents and their child.

Parent characteristics
Mothers (%)
Race (%)
White
Black
Other
Ethnicity (%)
Hispanic/latino
Not Hispanic/latino
Education level (%)
Less than BS degree
BS degree
Graduate degree
Family income (%)
<$27,564
$27,565$63,875
>$63,875
Employed (% yes)
Parent BMI (kg/m2)
Family dinner (%)
02 times/week
35 times/week
67 times/week

Boys
(n = 102)

Girls
(n = 108)

p-value

88

90

0.71
0.35

88
4
5

82
3
9

11
87

9
89

38
22
40

33
27
40

14
30
56
67
25

14
28
58
69
25

21
30
49

10
32
57

40

43

61
25
11

70
20
6

20
79
1
52

15
83
2
67

0.93

0.62

0.91

Child characteristics
Age (months)
Birth order (%)
Oldest child
Second oldest child
Third oldest child
Weight status (%)
Underweight
Healthy weight
Overweight
Perceived as good eaters (% yes)

0.67
0.89
0.11

0.21
0.41

0.58

3.3. Relations among food neophobia, parental feeding practices, and

child food preferences
0.03

Table 2
Participant food neophobia and parental feeding practices Scores.

Food neophobia scores

Parent Neophobia
(range 550)
Child Neophobia
(range 550)
Correlation between
parentchild
Food neophobia scores
(p-value)*
Parent eating history
Parent eating then
(range 525)
Parental feeding practices
Child Autonomy (range 15)
Pressure to eat (range 15)
Urging new foods (range 15)
Offering new foods
(range 15)

Boys
(n = 102)

Girls
(n = 108)

Total
(n = 210)

Mean

SD

Mean

SD

Mean

SD

23.0

8.3

20.8

29.4

21.9

7.4

31.5

9.3

29.4

8.2

30.4

8.8

r = 0.09

r = 0.17

r = 0.14

(p = 0.35)

(p = 0.08)

(p = 0.04)

16.0

16.7

2.8
2.7
3.2
3.4

significantly correlated between both parentson dyads (r = 0.45,

p < 0.001) and parentdaughter dyads (r = 0.24, p = 0.01). Gender
differences emerged within the concordance data, including the
number of foods parentson dyads reported as dislike and eat
(r = 0.39, p < 0.0001) and parentdaughter dyads reported as
dislike, do not eat (r = 0.29, p = 0.002). Children who were older
(p = 0.04), perceived by parents to be of a healthy weight
(p = 0.002) and good eaters (p < 0.0001) had the highest agreement with their parents food preferences.
When concordance data were analyzed by food groups (data
not shown), parentchild dyads had >75% concordance in preferences for whole grains, fruit, starchy vegetables, and dairy. Lowest
concordance (3966%) was noted for other vegetables (e.g., celery,
mushrooms, cucumber, and onions), sweetened beverages, and
entrees. Discordance occurred when parents had never offered
their child a food (e.g., vegetables and proteins), and was not influenced by demographic factors.
Parent and child food preference concordance was also examined for five selected food groups (Howard et al., 2012) (vegetables,
fruits, core foods [e.g., chicken, yogurt, eggs], non-core foods [e.g.,
pizza, French-fries, cookies], and sweet beverages [e.g., flavored
milk, soda, apple juice]). Each food group contained between 2
and 13 independent food items. Girls who were categorized as a
healthy weight and good eaters by their parents were reported
to have the highest concordance for foods with their parents on
both vegetables (p = 0.04) and core food items (p < 0.0001). The
highest concordance scores for sweetened beverages (p = 0.04)
were found among older Hispanic parent-son dyads.

4.0

0.9
0.87
0.5
0.6

3.0
2.7
3.1
3.4

3.7

0.9
0.87
0.5
0.5

16.4

2.9
2.7
3.2
3.4

3.8

0.9
0.87
0.5
0.5

*
Calculated using Spearman correlation coefficients at the p < 0.05 level of
significance.

Fig. 1 depicts the measured constructs along the hypothesized

pathway of a childs development of food preferences. Briefly, it
was hypothesized that parental feeding practices would be associated with child food neophobia and in turn with a childs overall
preference for foods. Adult food neophobia was not correlated with
parents eating history (r = 0.002, p = 0.99), suggesting that how
parents remembered they were fed as children was unrelated to
their self-reported ratings of their own food neophobia. Adult
self-ratings of food neophobia were negatively correlated with
Child Autonomy (r = 0.19, p = 0.006) and offering new foods
(r = 0.25, p = 0.003) revealing that parents who are more neophobic allow their child less autonomy in self-feeding and also offer
fewer new foods to their children. While not reaching statistical
significance, pressure to eat (r = 0.12, p = 0.07) and urging new
foods (r = 0.12, p = 0.09) were positively correlated with adult food
neophobia scores: parents who were more neophobic appear to
have used more controlling feeding practices to try to influence
their child to eat.
Child neophobia was significantly and positively correlated
with pressure to eat (r = 0.29, <0.0001) and urging of new foods
(r = 0.16, p = 0.02), while offering new foods (r = 0.40,
p < 0.0001) was found to be negatively correlated with child food
neophobia. Child food neophobia was moderately and negatively
correlated with overall number of the foods that a parent reported
their child liking and eating (e.g., food preferences) (r = 0.51,
p < 0.0001).

3.2. Concordance in food preferences between parentchild dyads

Parents classified the majority of foods (n = 40 foods rated) as
like and eat for both the child and adult (Table 3). The number
of foods that the child liked and ate was moderately correlated
with the number of foods that the parent liked and ate for both
boys (r = 0.32, p = 0.001) and girls (r = 0.25, p = 0.01). Similarly,
the number of foods categorized as like, do not eat were

3.4. Variables associated with the number of healthy foods children

liked and ate
To further understand the variables associated with children
liking and eating healthy foods (e.g., vegetables, fruits, and core
foods), a forward stepwise linear regression analyses was conducted
(separately for boys and girls) for each food group (Tables 4 and 5).

61

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

Table 3
Parent and child food preference concordance, mean (min, max).
Like and eat

Like, do not eat

Dislike and eat

Dislike, do not eat

Never tasted

Boys
Adult
Child
Correlation
(p-value)

31 (15,40)
29 (12,38)
0.32
(p = 0.001)

4 (0,21)
1(0,9)
0.45
(p < 0.0001)

1 (0,15)
1 (0,6)
0.39
(p < 0.0001)

3 (0,13)
5 (0,26)
0.15
(p = 0.14)

0 (0,2)
3 (0,12)
0.09
(p = 0.33)

Girls
Adult
Child
Correlation
(p-value)

31 (18,39)
25.5 (13,38)
0.25
(p = 0.01)

3 (0,14))
1 (0,11)
0.24
(p = 0.01)

1 (0,5)
1 (0,5)
0.04
(p = 0.70)

2 (0,12)
4 (0,18)
0.29
(p = 0.002)

0 (0,5)
3 (0,14)
0.18
(p = 0.06)

Table 4
Variables associated (final model) with the # of specific foods by food groups parents report their boys like and eat.
# of vegetables likes and
eats

# of fruits likes and

eats

# of core foods likes and

eats

# of non-core foods
likes and eats

# of sweet beverages likes and

eats

B (95% CI)
R2 = 0.53
effect size = 0.52

B (95% CI); p-value

R2 = 0.32
effect size = 0.29

B (95% CI); p-value

R2 = 0.03
effect size = 0.02

B (95% CI); p-value

R2 = 0.31
effect size = 0.24

B (95% CI); p-value

R2 = 0.12
effect size = 0.10

Race
White
Black
Hispanic
Education
Less than BS degree
BS degree
Graduate degree
Income
<$27,564
$27,565$63,875
>$63,875
Employment Status
(yes)
BMI
Parent food
neophobia
Parent eating history
Child Autonomy
PRESSURE TO EAT
Offering new foods
Urging new foods
Child age (months)
Birth order
Oldest child
Second oldest child
Third oldest child
Child weight status
Underweight
Healthy weight
Overweight
Child food neophobia

ref
1.75 ( 3.39,

0.11)*

ref
0.53 ( 0.06, 1.13)

0.04 ( 0.00, 0.09)*

0.32 (0.06, 0.58)*

0.51 (0.06, 0.96)*

0.19 ( 0.22,

0.15)***

ref
3.55 ( 5.99, 1.11)**
0.08 ( 0.11, 0.05)***

0.05 (0.02, 0.07)***

0.02 (0.01, 0.03)***

ref
1.02 (0.19, 1.85)

ref
0.16 ( 0.18, 0.51)

1.97 ( 2.87,
1.06)***
ref
0.06 ( 0.10,

0.02)**

Note: Forward stepwise regression models were used in this analysis and only significant variables were selected for the final model. Variables listed in the table that remain
blank represent variables tested in previous steps but did not remain significant for final model.
*
p < 0.05.
**
p < 0.01.
***
p < 0.001.

In general, child food neophobia was associated with liking and eating fewer food items in both sexes. In the final model, food neophobia was the strongest predictor for boys eating fewer vegetables
(b = 0.19, p < 0.001) and predicted 53% of the variance. Boys with
higher food neophobia scores also liked and ate significantly fewer
fruits (b = 0.08, p < 0.001). In boys, liking and eating more fruits
was significantly associated with lower child weight (b = 3.55,
p < 0.01) and greater Child Autonomy (b = 0.32, p < 0.05). Furthermore, boys who liked and ate higher numbers of non-core food items

had lower child food neophobia score (b = 0.06, p < 0.01), higher
weight status (b = 1.97, p < 0.001), and were more often pressured
to eat (b = 0.51, p < 0.05). For girls, liking and eating more vegetables
was significantly associated with higher reported pressure to eat
(b = 0.59, p < 0.01), increased age (b = 0.04; p < 0.01, and a lower
food neophobia score (b = 0.17; p < 0.001). The number of core
foods a girl liked and ate was predicted by both child food neophobia
(b = 0.10; p < 0.001) as well as how the parents eating history
(b = 0.09; p < 0.05).

62

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

Table 5
Variables associated (final model) with the # of specific foods by food groups parents report their girls like and eat.

Race
White
Black
Hispanic
Education
Less than BS degree
BS degree
Graduate degree
Income
<$27,564
$27,565 $63,875
>$63,875
Employment status (yes)
BMI
Parent food neophobia
Parent eating history
Child Autonomy
Pressure to eat
Offering new foods
Urging new foods
Child age (months)
Birth order
Oldest child
Second oldest child
Third oldest child
Child weight status
Underweight
Healthy weight
Overweight
Child food neophobia

# of vegetables likes and eats

# of core foods likes and eats

# of non-core foods likes and eats

# of sweet beverages likes and eats

B (95% CI)
R2 = 0.41
effect size = 0.38

B (95% CI); p-value

R2 = 0.23
effect size = 0.21

B (95% CI); p-value

R2 = 0.04
effect size = 0.02

B (95% CI); p-value

R2 = 0.04
effect size = 0.03

ref
2.26 ( 0.51, 5.03)

ref
0.42 ( 1.34, 0.50)

0.02 (0.00, 0.04)*

0.09 ( 0.18,

0.01)*

0.10 ( 0.14,

0.06)***

0.59 (0.21, 0.97)**

0.04 (0.01, 0.06)**

0.17 ( 0.21,

0.13)***

Note: No predictors were significantly associated with # of fruits therefore no final model conducted.
*
p < 0.05.
**
p < 0.01.
***
p<0.001.

4. Discussion
This study examined the relationships among parental feeding
practices and parent/child food neophobia with child food preferences. In this sample of 210 parentchild dyads, pressuring a child
to eat was found to be positively correlated with child food neophobia (i.e., increased childs food neophobia), while offering a child
new foods was found to be negatively correlated (i.e., decreased
childs food neophobia). Furthermore, a childs food neophobia
score was moderately correlated with the childs liking and eating
a variety of foods assessed in our food preferences instrument.
The literature has been consistent in reporting that high scores
for child food neophobia are predictive of a child liking and eating
fewer fruits and vegetables (Cooke et al., 2006; Howard et al.,
2012). In a study by Skinner et al., the strongest predictor of the
number of foods children liked at 8 years of age was their food neophobia score (Skinner et al., 2002). In this sample of parentchild
dyads, parents tended to offer their children foods they themselves
liked and ate and as a result, parents reported their child liking and
eating similar foods. This evidence is consistent with other studies
further illustrating the strong concordance of food preferences
between parentchild dyads (Cathey & Gaylord, 2004; Cooke
et al., 2004; Falciglia, Pabst, Couch, & Goody, 2004; Howard et al.,
2012; Skinner et al., 1998). In a sample examining Australian
mothers and their toddlers, maternal food preferences, specifically
liking, for vegetables, fruits, and non-core foods was positively
associated with their childs liking of similar foods (Howard
et al., 2012). Likewise, mothers who did not like vegetables
themselves, did not offer them to their children; therefore, the

children were never exposed to vegetables (Howard et al., 2012).

This may be compounded by parents in some cases projecting their
own food preferences on to the child (Mata, Scheibehenne, & Todd,
2008), which can limit the variety and healthfulness of foods
offered to the child. In the current study, our findings took this
relationship one step further by finding that mothers who reported
not eating vegetables themselves, but offered their child these
foods, reported that their children liked and ate vegetables, despite
the mothers own preference. It is very important that caregivers of
young children recognize the importance of modeling healthy eating habits by not only eating these healthy foods in front of their
children, but also by offering these foods to their children at a very
young age in a non-contingent manner.
To date, studies that have examined relationships between food
neophobia, parental feeding practices and food preferences have
been limited to examining single components of this complicated
relationship rather than simultaneously examining the variables
in one model. Furthermore, parents play a crucial role in what the
child eats. What a parent provides for food choices may be based
on the parents own preferences or food neophobia, in particular
with respect to healthy food choices (Tan & Holub, 2012). Hence,
relying on a parents report of their childs food neophobia may
underestimate
the
childs
actual
neophobia
(Laureati,
Bergamaschi, & Pagliarini, 2015) or may project parents own
behaviors onto their children (Mata et al., 2008). Studies have
examined the relationship between parental food preferences and
child food preferences (Howard et al., 2012; Morrison, Power,
Nicklas, & Hughes, 2013; Vereecken et al., 2010), parent and child
food neophobia (Faith et al., 2003; Skinner et al., 2002), or specific

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

parental feeding practices (e.g., pressuring) with a childs fruit/vegetable intake (Gregory et al., 2011; Vereecken et al., 2010). Others
have examined the relationship between a childs food neophobia
and their food preference (Galloway et al., 2003; Howard et al.,
2012). The current study adds to the existing literature by constructing and testing a theoretical model of the development of
food preferences using many of the previously explored relations.
By using such a model, each variable could concurrently be examined within this framework. This study highlights the possible contributing factors of two parental feeding strategies, pressuring and/
or urging a child to eat, and how these practices are negatively associated with childs food neophobia. The findings related to offering
children new foods suggests that parents have the potential to facilitate the childs willingness to try and learn to like new foods.
Explicitly identifying potential positive outcomes, such as moderating a childs food neophobia and food refusal, are essential to help
caregivers make informed decisions about strategies to employ
when teaching their children how to eat healthfully. Parents good
intentions may, in fact, be hindering their success.
This study also examined the association of parents food neophobia scores with their feeding practices and revealed that parents who are less food neophobic themselves tend to give their
children more eating autonomy and offer their child new foods.
Parents who have less fear of trying new foods may provide a more
positive feeding environment at home for their children with more
constructive mealtime interactions that positively influenced food
acceptance. These key findings may be helpful for intervention
planning as improving parental awareness that their own eating
characteristics may play an important role in influencing their
childs eating and the decisions parents make about how to teach
their children to eat.
In our sample, parents with higher food neophobia scores were
less likely to give their child control regarding self-feeding (i.e.,
autonomy in feeding). The combination of parental neophobia,
poor food choices in the home environment, and lack of autonomy
in eating may negatively impact the childs development of healthy
eating habits. As a parent, it may be useful to know that their decisions on how to feed their child has the potential to improve their
overall goal of having a child that likes and eats a variety of healthy
foods. Unlike the practice of pressuring a child to eat, simply offering children new foods in a non-coercive manner that gives children more control could decrease childrens reluctance to try
foods. In contrast, urging a child to eat (which can be thought of
as more intensively offering a food or making the access to other
more desired foods contingent on trying the new food) may have
the opposite effect.
Although this study was able to examine the relations among
parental feeding practices, food neophobia and food preferences
in a sample of parents with preschool aged children, it is not without limitations. First, parents who participated in this study were
employed at one of two universities targeted for recruitment.
Although the University campuses represented different academic
settings and a diversity of employees, this method of recruitment
does have the potential to bias our results by including the experiences of mostly middle-income to high income non-Hispanic families. The majority of respondents identified themselves as
mothers hence these findings would need to be confirmed in a
more equal sample of mothers and fathers as we know that
research in the area of fathers feeding practices may have its
own unique findings (Harris & Ramsey, 2015; Khandpur, Blaine,
Fisher, & Davison, 2014; Vollmer, Adamsons, Foster, & Mobley,
2015). Therefore, our results may not be generalizable to cohorts
examining fathers feeding practices or into lower income families
or families from other ethnicities and cultures. Second, the data for
this analysis were collected using parental report of their childs
eating behaviors. Parents may have misreported some of their

63

feeding practices or reported that their children liked and ate more
of the healthier food items in an effort to conform with social
norms or with their own goal for their children to be good eaters.
Lastly, there are many factors that may affect the development of
child food preferences that we did not include in our model. The
ideal study would include all of the factors known to be associated
with food preference development and then most accurately predict using a longitudinal design which factors are most influential.
To date the majority of studies examining the influences on the
development of a childs food preferences including this study have
been cross-sectional. Few studies have investigated the causal
pathway by using appropriate longitudinal design and pathway
modeling, therefore future research should focus on such a design.
Longitudinal research has the potential to elucidate which traits
(e.g., parent and/or child food neophobia) and parental feeding
practices positively contribute to the development of a childs preference for certain foods. Such studies could inform the development of best practice guidelines which can assist parents in
adopting feeding strategies that help children eat a healthy diet.
Furthermore, a pathway modeling approach or Structural Equation
Modeling (SEM) would allow statistical testing of the temporally
occurring factors along the theoretical pathway between parental
feeding behaviors and child food preferences.
In future studies a more objective measure, such as conducting
home visits to video record a parentchild meal interaction, need
to be conducted to confirm these findings. Once researchers understand what role parental feeding practices and food neophobia
have in the development of childrens food preferences, interventions can be created to guide parents toward the best strategy
for having a child that happily eats his or her vegetables.
Source of funding
NIH/NCRR Colorado CTSI Grant No. UL1 TR000154.
Contributions
JLK formulated the research question and wrote the first draft of
the manuscript. JLK and ALBS conducted the analysis. SLJ designed
the study and carried out the study to completion. JLK, ALBS, DMF,
SLJ contributed to the writing of the manuscript. All authors
reviewed and approved the final draft prior to submission.
Conflict of interest
The authors declare that they have no conflicts of interest.
Acknowledgements
We would like to thank all of our participants for taking the
time to complete our survey and participating in our research.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.foodqual.2016.01.
006.
References
Birch, L. L. (1999). Development of food preferences. Annual Review of Nutrition, 19,
4162.
Birch, L. L., Fisher, J. O., Grimm-Thomas, K., Markey, C. N., Sawyer, R., & Johnson, S. L.
(2001). Confirmatory factor analysis of the Child feeding questionnaire: A
measure of parental attitudes, beliefs and practices about child feeding and
obesity proneness. Appetite, 36, 201210.

64

J.L. Kaar et al. / Food Quality and Preference 50 (2016) 5764

Birch, L. L., & Marlin, D. W. (1982). I dont like it; I never tried it: Effects of exposure
on two-year-old childrens food preferences. Appetite, 3, 353360.
Boles, R. E., Burdell, A., Johnson, S. L., Gavin, W. J., Davies, P. L., & Bellows, L. L. (2014).
Home food and activity assessment. Development and validation of an
instrument for diverse families of young children. Appetite, 80, 2327.
Boles, R. E., Reiter-Purtill, J., & Zeller, M. H. (2013). Persistently obese youth:
Interactions between parenting styles and feeding practices with child
temperament. Clinical Pediatrics (Phila), 52, 10981106.
Bova, A., & Arcidiacono, F. (2014). You must eat the salad because it is nutritious".
Argumentative strategies adopted by parents and children in food-related
discussions at mealtimes. Appetite, 73, 8194.
Branen, L., & Fletcher, J. (1999). Comparison of college students current eating
habits and recollections of their childhood food practices. Journal of Nutrition
Education, 31, 304310.
Brown, A., & Lee, M. (2011). Maternal child-feeding style during the weaning
period: association with infant weight and maternal eating style. Eating
Behaviors, 12, 108111.
Cassells, E. L., Magarey, A. M., Daniels, L. A., & Mallan, K. M. (2014). The influence of
maternal infant feeding practices and beliefs on the expression of food
neophobia in toddlers. Appetite, 82, 3642.
Cathey, M., & Gaylord, N. (2004). Picky eating: A toddlers continuing approach to
mealtime. Pediatric Nursing, 30, 101107.
Cooke, L., Carnell, S., & Wardle, J. (2006). Food neophobia and mealtime food
consumption in 45 year old children. International Journal of Behavioral
Nutrition and Physical Activity, 3, 14.
Cooke, L. J., Haworth, C. M., & Wardle, J. (2007). Genetic and environmental
influences on childrens food neophobia. American Journal of Clinical Nutrition,
86, 428433.
Cooke, L. J., & Wardle, J. (2005). Age and gender differences in childrens food
preferences. British Journal of Nutrition, 93, 741746.
Cooke, L. J., Wardle, J., Gibson, E. L., Sapochnik, M., Sheiham, A., & Lawson, M. (2004).
Demographic, familial and trait predictors of fruit and vegetable consumption
by pre-school children. Public Health Nutrition, 7, 295302.
Dovey, T. M., Staples, P. A., Gibson, E. L., & Halford, J. C. (2008). Food neophobia and
picky/fussy eating in children: A review. Appetite, 50, 181193.
Faith, M. S., Heo, M., Keller, K. L., & Pietrobelli, A. (2013). Child food neophobia is
heritable, associated with less compliant eating, and moderates familial
resemblance for BMI. Obesity (Silver Spring), 21, 16501655.
Faith, M. S., Heshka, S., Keller, K. L., Sherry, B., Matz, P. E., Pietrobelli, A., et al. (2003).
Maternal-child feeding patterns and child body weight: findings from a
population-based sample. Archives of Pediatrics and Adolescent Medicine, 157,
926932.
Falciglia, G., Pabst, S., Couch, S., & Goody, C. (2004). Impact of parental food choices
on child food neophobia. Childrens Health Care, 33, 217225.
Fletcher, J., Branen, L. J., & Lawrence, A. (1997). Late adolescents perceptions of their
caregivers feeding styles and practices and those they will use with their own
children. Adolescence, 32, 287298.
Galloway, A. T., Lee, Y., & Birch, L. L. (2003). Predictors and consequences of food
neophobia and pickiness in young girls. Journal of the American Dietetic
Association, 103, 692698.
Gregory, J. E., Paxton, S. J., & Brozovic, A. M. (2010). Maternal feeding practices, child
eating behaviour and body mass index in preschool-aged children: A
prospective analysis. International Journal of Behavioral Nutrition and Physical
Activity, 7, 55.
Gregory, J. E., Paxton, S. J., & Brozovic, A. M. (2011). Maternal feeding practices
predict fruit and vegetable consumption in young children. Results of a 12month longitudinal study. Appetite, 57, 167172.
Harris, T. S., & Ramsey, M. (2015). Paternal modeling, household availability, and
paternal intake as predictors of fruit, vegetable, and sweetened beverage
consumption among African American children. Appetite, 85, 171177.
Harris, P. A., Taylor, R., Thielke, R., Payne, J., Gonzalez, N., & Conde, J. G. (2009).
Research electronic data capture (REDCap)a metadata-driven methodology
and workflow process for providing translational research informatics support.
Journal of Biomedical Informatics, 42, 377381.

Haszard, J. J., Williams, S. M., Dawson, A. M., Skidmore, P. M., & Taylor, R. W. (2013).
Factor analysis of the comprehensive feeding practices questionnaire in a large
sample of children. Appetite, 62, 110118.
Howard, A. J., Mallan, K. M., Byrne, R., Magarey, A., & Daniels, L. A. (2012). Toddlers
food preferences. The impact of novel food exposure, maternal preferences and
food neophobia. Appetite, 59, 818825.
Johnson, S. L., Davies, P. L., Boles, R. E., Gavin, W. J., & Bellows, L. L. (2015). Young
childrens food neophobia characteristics and sensory behaviors are related to
their food intake. The Journal of nutrition.
Khandpur, N., Blaine, R. E., Fisher, J. O., & Davison, K. K. (2014). Fathers child feeding
practices: a review of the evidence. Appetite, 78, 110121.
Laureati, M., Bergamaschi, V., & Pagliarini, E. (2015). Assessing childhood food
neophobia: Validation of a scale in Italian primary school children. Food Quality
and Preference, 40(Part A), 815.
Laureati, M., Bertoli, S., Bergamaschi, V., Leone, A., Lewandowski, L., Giussani, B.,
et al. (2015). Food neophobia and liking for fruits and vegetables are not related
to Italian childrens overweight. Food Quality and Preference, 40(Part A),
125131.
Mata, J., Scheibehenne, B., & Todd, P. M. (2008). Predicting childrens meal
preferences: how much do parents know? Appetite, 50, 367375.
Mikkila, V., Rasanen, L., Raitakari, O. T., Pietinen, P., & Viikari, J. (2005). Consistent
dietary patterns identified from childhood to adulthood: The cardiovascular risk
in Young Finns Study. British Journal of Nutrition, 93, 923931.
Morrison, H., Power, T. G., Nicklas, T., & Hughes, S. O. (2013). Exploring the effects of
maternal eating patterns on maternal feeding and child eating. Appetite, 63,
7783.
Mustonen, S., Oerlemans, P., & Tuorila, H. (2012). Familiarity with and affective
responses to foods in 811-year-old children. The role of food neophobia and
parental education. Appetite, 58, 777780.
Newman, J., & Taylor, A. (1992). Effect of a means-end contingency on young
childrens food preferences. Journal of Experimental Child Psychology, 53,
200216.
Northstone, K., & Emmett, P. M. (2008). Are dietary patterns stable throughout early
and mid-childhood? A birth cohort study. British Journal of Nutrition, 100,
10691076.
Pliner, P. (1994). Development of measures of food neophobia in children. Appetite,
23, 147163.
Pliner, P., & Hobden, K. (1992). Development of a scale to measure the trait of food
neophobia in humans. Appetite, 19, 105120.
Russell, C. G., & Worsley, A. (2008). A population-based study of preschoolers food
neophobia and its associations with food preferences. Journal of Nutrition
Education and Behavior, 40, 1119.
Scaglioni, S., Salvioni, M., & Galimberti, C. (2008). Influence of parental attitudes in
the development of children eating behaviour. British Journal of Nutrition, 99
(Suppl. 1), S22S25.
Skinner, J. D., Carruth, B. R., Wendy, B., & Ziegler, P. J. (2002). Childrens food
preferences: A longitudinal analysis. Journal of the American Dietetic Association,
102, 16381647.
Skinner, J., Ruth, Carruth B., Moran, Iii J., Houck, K., Schmidhammer, J., Reed, A., et al.
(1998). Toddlers food preferences: Concordance with family members
preferences. Journal of Nutrition Education, 30, 1722.
Tan, C. C., & Holub, S. C. (2012). Maternal feeding practices associated with food
neophobia. Appetite, 59, 483487.
Ventura, A. K., & Birch, L. L. (2008). Does parenting affect childrens eating and
weight status? International Journal of Behavioral Nutrition and Physical Activity,
5, 15.
Ventura, A. K., & Worobey, J. (2013). Early influences on the development of food
preferences. Current Biology, 23, R401408.
Vereecken, C., Rovner, A., & Maes, L. (2010). Associations of parenting styles,
parental feeding practices and child characteristics with young childrens fruit
and vegetable consumption. Appetite, 55, 589596.
Vollmer, R. L., Adamsons, K., Foster, J. S., & Mobley, A. R. (2015). Association of
fathers feeding practices and feeding style on preschool age childrens diet
quality, eating behavior and body mass index. Appetite, 89, 274281.